Check out my HOMster thread. In the thread I measure an extremely undersized horn, with and without the HOM reducing foam. In the measurements you can see that the foam is reducing efficiency to a significant degree - as much as 10-15dB.
I believe this demonstrates that a significant portion of the energy is due to HOMs.
Recently I've also measured an OS-style waveguide, and will post those results shortly. The OS waveguide had a much smaller degree of attenuation from the foam.
It is my hypothesis that the OS waveguide suffers from HOMs to a lesser degree, and that is why the foam has less attenuation in an OS waveguide than in a horn.
Stay tuned for measurements...
Seems to me that this difference isn't worth worrying about. When I asked B&C they had to look it up because no one had asked before. (The answer was 6.5 degrees.) Now we are worrying about a difference of less than a degree? I'll bet that if you ask again you'll get yet a different number. And do you realize what kind of tollerence you need to hold to 1 degree? It probably varies between 6.5 and 7.3 degrees in production!!!
Let's not make yet another mountain out of a mole hill.
An interesting hypothesis. I know in my situation its only about 2-3 dB loss. There are lots of densities of foam and the loss is highly dependent on the density and there are even differences across manufacturers. That I know to be true from many measurements. So first and formost you need to be certain than the foam itself is not the biggest variable. And as I recal, you do not use a solid piece, but layers. This too could also be a factor.
There are lots of potential variables here and we should be careful in our experiments to control them.
complexity by degrees
Not trying to...just want to know where the "bullseye" is. ...so, I'll make a plug/gage and match the DE250's I have exactly...I guess I don't need a number, or any math...
...and that was not an attempt to put words into your mouth...it was phrased as a question... It was an attempt of my simple mind to wrap itself around the concept...
Last edited:
I'm trying to build a pair of nearfield monitors using ScanSpeak 6600 tweeters. (You honestly don't need that much SPL if you're only a meter from the speakers.) While these tweeters are supposed to have no trouble with a 1.7khz xover point, I'd like to reduce distortion and power compression by using a waveguide to increase efficiency in the lower midrange. Considering that I'm crossing over to a Seas Excel 7" midwoofer in a small sealed box (I've got a sub I'll be crossing over at 70hz), what should I do for optimal performance? I'm open to having someone CNC me some waveguides out of polyethylene or MDF.
...and that was not an attempt to put words into your mouth...it was phrased as a question... It was an attempt of my simple mind to wrap itself around the concept...
HOMs tend to be low in amplitude compared to the primary wave, or at least that what I try and do. But at certain frequencies they can be very efficient and so even the low amplitudes can make the radiated sound significant. This does not imply that they are greater in level than the primary wave, thats not likeley but COULD happen, nor that they are audible, although the greater the level the greater the likelihood that they are audible.
Hope this helps and sorry if I jumped on you unintentionally.
All that I would say is to use a compression driver and a standard waveguide regardless of being nearfield or not. Its a considerable complication to any design to add a waveguide to a direct radiating device, unless its a broad waveguide, which doesn't change things very much, but also doesn't improve things very much.
Based on your feedback, I re-ran all the measurements, and they were consistent with what you found. In a waveguide that is basically oblate spheroidal*, the attenuation was about 2-3dB at the top end.
On the other hand, the second set of measurements was consistent with what I found over the summer - in a horn with a very narrow coverage angle, the loss at high frequencies was about 12dB!
Again, I think this suggest that bad horns have more HOMs.
Admittedly, I'm not using a solid bun of reticulated foam, but I *am* using the same type of foam for both devices.
As I understand it, HOMs cannot be generated if the dimensions are less than a wavelength. The horn under test has a height of approximately two inches. A wavelength of 6500hz is two inches long. This may play a factor too. In other words, the small dimensions of the horn may lead to an outrageous level of HOMs above 6500hz, because the horn can only generate HOMs on one axis below that frequency.
I'll be posting measurements soon, but I'm flying to NYC for work shortly.
Last edited:
Look forward to seeing them as I have a quad of that WG here to be used shortly. Been too busy with work and studies to measure them yet. I'll be using my DE250's.
Based on your feedback, I re-ran all the measurements, and they were consistent with what you found. In a waveguide that is basically oblate spheroidal*, the attenuation was about 2-3dB at the top end.
{snip}
I'll be posting measurements soon, but I'm flying to NYC for work shortly.
* here's the waveguide in question. Swiped the pic off Brandon's site.
Okay, from reading some more over the weekend and making a spreadsheet from the book, I think you're talking about their 'predicted in room response' curve? And yes, it seems that this would be most helpful to know since a lot of the conclusions later in the book depend on it...
Wouldn't high directivity speakers still be fairly / accurately represented using their metrics? There would just be less energy in the reflections that they average together. Is what you're talking about (and what I'd tend to say) that they then used these metrics to promote low directivity designs as being more 'preferable' to other speakers? But also that 'preferable' is not necessarily 'accurate'? Would it be better to say that low directivity designs are easier to get wider apparent source width from in a typical setup which some people find preferable? What would you say is the advantage of a higher directivity design in this context - more pinpoint imaging? Or that by having wide enough constant directivity and aiming the speakers in as you talk about you can get the opposite wall reflection to have a high enough level that you can still get image broadening?
* here's the waveguide in question. Swiped the pic off Brandon's site.
What waveguide is that and where can you purchase it?
Hey John
I'd say that you have quite a good grasp on the issues, there's almost nothing that I can add, but maybe some discussion will clarify some points for others.
Yes, I do believe that the Harman work would tend to find low directivity designs "preferable" to or comparable with higher directivity designs owing to several assumptions and techniques buried in their work. "Preference" is not necessarily "accuracy" and many people have pointed this out.
I am writting a white paper right now on what the difference between low and high directivity loudspeaker systems will be in a small room. What it means subjectively regarding "image" is somewhat conjecture, but it is easy to show that a higher directivity will have fewer confounding early reflections than a low directivity speaker. This is most pronounced in a small room - as the room gets larger the differences fade and in a very large room the subjective difference due to directivity will vanish. In a large room, directivity is used to get the sound where it is most needed, to not waste precious power. These requirements for directivity are completely different and it is most curious that high directivity would be desirable in both sized rooms but for completely different reasons. Many people discount high directivity in small rooms because they understand the use of directivity in a large room and do not see the same need in a small room. But this ignores the fact that the rational for high directivity in these two rooms is completely different.
The Harman test rooms are all larger than my listening room and, I would say, tend to be on the large side.
What is clear to me and not at all contridictory to Floyds work, is that two things are a benefit in a small room. First good sound quality requires many late reflections to add a real room ambiance or spaciousness, but there is more than one way to do this. Where I differ with Floyd is that I want to see a fairly long dead space between the direct sound and this reverberant tail so as not to confound our sense of image in the recording, and/or add coloration, with a lot of early reflections. Wide directivity will yield spaciousness in almost any room because of the large amount of early reflections and this is what Floyd finds, and I don't disagree.
But what I propose is to use high directivity to yield a larger dead zone after the direct sound and a very live rear part of the room to yield a long reverberation tail for good spaciousness. I believe that this is the preferable way to go and one that Floyd never tested. Its one that most people who hear my room agree is a good combination. I have a very small listening room, but the subjective effect is one of a much much larger space. This is created by the large dead zone followed by the large reverberation.
So I don't disagree with anything that Floyd did or said, only that he excluded what I propose as a prefered solution and hence we cannot say how it would compare. To me what we know about hearing supports my approach and what we know about marketing supports Floyds approach. Not at all surprising I would say. My approach IS NOT user friendly. Most rooms are not made like I suggest and neither are most speakers. If I were selling speakers for the "average" living room, I might find that Floyd was completely correct and that I'd sell more speakers if I made them like Harmans. My designs are very hard to make and costly. The room design is also difficult.
On the other hand, its starting to appear that my approach does indeed work well in any room - as my customers are saying. I know that it works very well in the rooms that I have designed and built.
Last edited:
Its found on the QSC HPR122i speaker.
We can buy them through the online QSC parts ordering website in the US but Im not sure where you can order them over there? You might want to contact QSC.
The small Compression driver that goes with it Celestion 1425, measured pretty good on Brandon's sit too and well worth its sub $70 price tag online.
Awesome!! Will that be in your HOM thread?
I have those waveguides and I have the retangular QSC waveguides found in the QSC HPR-152i.
There are some HUGE bargains on the QSC parts site but QSC has warned some of the DIY crowd that we shouldnt abuse the parts site and order too many
Hi Earl
Basically you're just recreating the acoustics of the average reflection-free-zone control room. That's the right thing to do because the original IS what the mixing or mastering engineer heard. He adjusted the spatial attributes of a recording within the acoustics of his specific control room to make it sound "right". So any room with similar acoustics or any setup that creates similar acoustics, will allow for a more accurate sound reproduction. It might not sound like "the real thing" but that's not upon us to decide - it's solely the decision of the mixing engineer.
Best, Markus
Basically you're just recreating the acoustics of the average reflection-free-zone control room. That's the right thing to do because the original IS what the mixing or mastering engineer heard. He adjusted the spatial attributes of a recording within the acoustics of his specific control room to make it sound "right". So any room with similar acoustics or any setup that creates similar acoustics, will allow for a more accurate sound reproduction. It might not sound like "the real thing" but that's not upon us to decide - it's solely the decision of the mixing engineer.
Best, Markus
Hey Markus
Good point (as usual). I had not thought of it in that context, but it does make sense. I did it because of what we know of hearing, but thats probably the same rational used much earlier in the RFZ concepts, so it all gels together.
Honestly, I started to doubt that there's a rational behind RFZ designs. Sound reproduction the way we know it might just be a cultural phenomenon closely tight to the spaces we live in, like church music is tight to the space a church provides or symphonic music to the classic concert hall.
Best, Markus
Best, Markus
Earl,
I agree that minimizing early reflections is the right way to do it.
What do you think about surround processing and rear speakers as an alternative to a live back end of the room?
With surround processing that extracts ambiance from the recording, you could make almost the same argument as for the fronts, i.e., let the ambiance come from the recording rather than the room.
I agree that minimizing early reflections is the right way to do it.
What do you think about surround processing and rear speakers as an alternative to a live back end of the room?
With surround processing that extracts ambiance from the recording, you could make almost the same argument as for the fronts, i.e., let the ambiance come from the recording rather than the room.